Free Our Books

because books want freedom, too

Why Gold OA is better than Green OA

with 7 comments

Imagine if there was a kind of free software called Green Free Software.  Perhaps would be accessible only to people who paid to subscribe to it.  Authors of patches for the Linux kernel would have to submit them first to Linus Torvalds; if the patch was accepted, the author could then self-archive it in a separate OA source-code repository after an embargo of one year.  There would be many different OA repositories of Linux patches, all containing different patches.  Companies like IBM and Intel, whose employees contribute to Linux, would have OA ‘company repositories’ for their employees’ patches.  Many authors wouldn’t bother doing the extra work to self-archive their patches.  As a result, it would be impossible to get all the latest patches (and hence to get a complete, up-to-date Linux kernel) without paying for a subscription to  In an attempt to solve this problem, some companies would establish ‘mandates’ to require their employees to self-archive in their company’s repsoitory.  Of course these mandates would not affect independent programmers.  Even if the mandates were 100% successful, anyone wishing to assemble a complete kernel would have to download patches from hundreds of different company repositories and combine them together.

Now ask yourself: would that be a better system than the one we have now, where anyone can download the latest complete kernel from for free?


Written by Benjamin Geer

26 July 2009 at 10:56

Posted in Opinion

Tagged with , ,

7 Responses

Subscribe to comments with RSS.

  1. See: “On the Deep Disanalogy Between Text and Software and Between Text and Data Insofar as Free/Open Access is Concerned”


    Stevan Harnad

    27 July 2009 at 17:40

    • Stevan, i read many of your texts/email on the differences between academic knowledge production and free and open source software. And i think you’re mistaken on so many accounts, and so set in your mistaken ways of thinking about this, that a blog comment is not a place to argue with you. However, i’d like to try to show you, using the same method you used in your The Giveaway/NonGiveaway Distinction text, a set of comparative slides, that similarities are many, on multiple levels. So, can you please spare me the time of recreating your green slides from that text from scratch, and email me your original slides (, so that i can show you how does such comparison looks like from my, and Ben’s, standpoint.

      Toni Prug

      27 July 2009 at 19:57

  2. Stevan, I think you misunderestimate the similarities between academic papers and software source code. As someone who has written a fair amount of both, I can tell you that in practice they are very similar in terms of production, distribution and use. Both are texts, which is why they’re both covered by copyright law. People write these texts, and on a large software project like the Linux kernel, they submit them to other people, who have recognised expertise and authority, for checking, correction and approval. This is very much like peer review, in that it is a collaborative process (there is often more than one reviewer), and often results in revision and improvement of the original submission. Once the text is accepted, it is placed in a repository where anyone can download it.

    You seem to be under the impression that free software is typically distributed by sending people copies; this is not the case. As with OA articles, it’s typically distributed by sending people links to the relevant repository so they can download it themselves. But without the fundamental freedom to copy the text, there could be no such repositories and therefore no such links. (A link is useless unless I have access to the repository, and downloading is a form of copying.)

    It’s true that some of the ways that people use software have no counterpart in the world of acacdemic writing. Yet one very important use is identical: people study software source code in order to learn how to write better programs, just as they study academic articles in order to learn how to write better articles. I myself learned programming this way, as have many others. One of the main goals of the free software movement was to protect this pedagogical use of software source code, which non-free software production threatened to make impossible.

    So the question is how to organise this type of process in the most efficient way, hence my post above.

    Benjamin Geer

    27 July 2009 at 18:26

  3. Ben, I think you’ve humpty-dumpterstood my point: Software code needs more re-use rights, but peer-reviewed research article texts do not. In “Why Gold OA is better than Green OA” you made an argument about software, not about OA (which is about and for article texts). Not only do researchers (unlike Wikipedia article “authors”) not do with one another’s article texts what programmers do with one another’s programmes, but they would not want it done. Software access, re-use and update issues have next to nothing to do with article access issues.

    (Research data access and re-use issues are something else again, books are yet another thing, and music/video/multimedia still another. It only generate confusion to conflate them.)

    I discuss the importance of these distinctions in this .mov file I made for the P2P meeting in Manchster in March:

    Stevan Harnad

    27 July 2009 at 19:44

  4. I agree that people do things with software that nobody ought to do with research articles, but I think that doesn’t undermine my argument, which is about a more basic question: given a set of texts that must be written and peer-reviewed, then read by a large audience, what is the most efficient system for providing free access to the texts? My argument is simply that it is simpler, more efficient and more reliable if the publishers themselves put the texts on centralised repositories where anyone can download them free of charge, rather than requiring the authors to do this. The simple reason is that if a task is centralised and happens automatically, it is more likely to happen consistently; if it requires the conscious intervention of innumerable individuals, it will be less efficient and more error-prone, e.g. because of the extra layers of administration involved in maintaining redundant repositories for the same texts, and in establishing and enforcing self-archiving mandates.

    Benjamin Geer

    27 July 2009 at 21:17

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: